Laura Schulz: The surprisingly logical minds of babies

翻译人员: Hong Li 校对人员: 杏儀 歐陽

Mark Twain summed up what I take to be

马克·吐温说过一句话，

one of the fundamental problems of cognitive science

在我看来，指出了认知科学

with a single witticism.

的根本问题。

他说，“科学非常奇妙，

He said, "There's something fascinating about science.

One gets such wholesale returns of conjecture

你实际上只需进行少量投资，

out of such a trifling investment in fact."

得到的回报却是一整套理论。”

(Laughter)

（笑声）

Twain meant it as a joke, of course, but he's right:

吐温当然是在开玩笑，但他没说错：

There's something fascinating about science.

科学就是这么神奇。

From a few bones, we infer the existence of dinosuars.

从几块骨头， 我们能推测出恐龙的存在。

From spectral lines, the composition of nebulae.

从几条光谱带， 我们能推测星云的构成物质。

From fruit flies,

分析果蝇，

the mechanisms of heredity,

我们能推导出遗传机制，

and from reconstructed images of blood flowing through the brain,

分析大脑血液流动的图像，

or in my case, from the behavior of very young children,

或者，从我的研究方向来说， 分析儿童的行为，

we try to say something about the fundamental mechanisms

我们尝试搞清楚人类认知的

of human cognition.

基本机制。

In particular, in my lab in the Department of Brain and Cognitive Sciences at MIT,

尤其在我们麻省理工学院 大脑和认知科学系实验室，

I have spent the past decade trying to understand the mystery

过去十年我一直在研究一个问题，

of how children learn so much from so little so quickly.

为什么小孩子能从无到有 快速地学会很多东西。

Because, it turns out that the fascinating thing about science

因为，科学的奇妙之处，

is also a fascinating thing about children,

恰恰也是小孩子的奇妙之处，

which, to put a gentler spin on Mark Twain,

从马克·吐温的话引申出来，

is precisely their ability to draw rich, abstract inferences

准确地说，就是他们都能 从少量的、充满干扰的数据中

rapidly and accurately from sparse, noisy data.

迅速而准确地得出丰富的理论推断。

I'm going to give you just two examples today.

我今天只举两个例子。

One is about a problem of generalization,

一个关于归纳总结，

and the other is about a problem of causal reasoning.

另一个关于因果推理。

And although I'm going to talk about work in my lab,

尽管我今天要谈的 是我的实验室里的工作，

this work is inspired by and indebted to a field.

但它的灵感来源于 整个（认知科学）领域。

I'm grateful to mentors, colleagues, and collaborators around the world.

我要感谢世界各地的 导师、同事和合作者们。

Let me start with the problem of generalization.

我先从归纳总结开始讲起。

Generalizing from small samples of data is the bread and butter of science.

从少量的数据样本进行归纳总结 是科学的立身之本。

We poll a tiny fraction of the electorate

我们调查一小部分选民的投票结果，

and we predict the outcome of national elections.

就能推测出大选结果。

We see how a handful of patients responds to treatment in a clinical trial,

我们分析临床试验中一部分病人 对治疗方案的反应，

and we bring drugs to a national market.

然后向全国市场推广新药。

But this only works if our sample is randomly drawn from the population.

但这要求我们抽取样本 要完全随机。

If our sample is cherry-picked in some way --

如果样本是刻意挑选的，

say, we poll only urban voters,

比如说，只抽取城市选民，

or say, in our clinical trials for treatments for heart disease,

或者，在治疗心脏病的临床试验中，

we include only men --

只抽取男性患者，

the results may not generalize to the broader population.

那结果可能不适用于整个人群。

So scientists care whether evidence is randomly sampled or not,

因此科学家非常重视 样本的抽取是否随机，

but what does that have to do with babies?

那婴儿会不会重视呢？

Well, babies have to generalize from small samples of data all the time.

实际上，婴儿一直在对 少量数据样本进行归纳总结。

They see a few rubber ducks and learn that they float,

他们见过几只橡胶鸭子， 知道它们能浮起来，

or a few balls and learn that they bounce.

见过几个球，知道它们能在地上弹跳。

And they develop expectations about ducks and balls

他们对鸭子和球产生了预判

that they're going to extend to rubber ducks and balls

并会在今后的人生中 将这种预判延伸到

for the rest of their lives.

（所有）橡胶鸭子和球身上。

And the kinds of generalizations babies have to make about ducks and balls

这种针对鸭子和球的归纳总结法，

they have to make about almost everything:

婴儿几乎要用在所有东西上：

shoes and ships and sealing wax and cabbages and kings.

鞋子、船、封蜡、卷心菜和国王。

So do babies care whether the tiny bit of evidence they see

那么婴儿会不会在乎 他们看到的这几个样本

is plausibly representative of a larger population?

是不是具有代表性呢？

Let's find out.

我们来看一看。

I'm going to show you two movies,

我将给你们放两段视频，

one from each of two conditions of an experiment,

每一段各反映一个实验里的一种情况，

and because you're going to see just two movies,

因为只有两段视频，

you're going to see just two babies,

所以你们只能看到两个婴儿，

and any two babies differ from each other in innumerable ways.

而任意两个婴儿之间都是千差万别的。

But these babies, of course, here stand in for groups of babies,

当然，这两个婴儿， 各代表一类婴儿，

and the differences you're going to see

你们即将看到的差别，

represent average group differences in babies' behavior across conditions.

代表了婴儿在不同情况下 普遍的行为差异。

In each movie, you're going to see a baby doing maybe

在每段视频中，婴儿的所作所为，

just exactly what you might expect a baby to do,

可能会跟你所预期的一样，

and we can hardly make babies more magical than they already are.

婴儿是如此神奇， 可能超乎你的想象。

但在我看来神奇的是，

But to my mind the magical thing,

and what I want you to pay attention to,

我也希望大家能注意到，

is the contrast between these two conditions,

就是两种情况之间的差别，

because the only thing that differs between these two movies

因为两段视频唯一的不同之处

is the statistical evidence the babies are going to observe.

就是婴儿需要观察的统计学证据。

We're going to show babies a box of blue and yellow balls,

我们会给婴儿看一个盒子， 里面装满了蓝色和黄色的球，

and my then-graduate student, now colleague at Stanford, Hyowon Gweon,

我当时的研究生学生， 现在是斯坦福大学的同事，权孝媛。

is going to pull three blue balls in a row out of this box,

会从盒子里连续拿出三个蓝色的球，

and when she pulls those balls out, she's going to squeeze them,

当她把球拿出来的时候，她会捏它们，

and the balls are going to squeak.

球会发出声音。

And if you're a baby, that's like a TED Talk.

对孩子来说，这就像TED演讲。

It doesn't get better than that.

真的没什么区别。

(Laughter)

（笑声）

But the important point is it's really easy to pull three blue balls in a row

重要的一点是， 从一个几乎全都是蓝色球的盒子里，

out of a box of mostly blue balls.

连续拿出三个蓝色的球非常容易。

You could do that with your eyes closed.

闭上眼睛都能做到。

It's plausibly a random sample from this population.

这是一个真正的随机取样。

And if you can reach into a box at random and pull out things that squeak,

如果你从一个盒子里随机 取出来的东西能捏响，

then maybe everything in the box squeaks.

那也许这个盒子里 所有的东西都能捏响。

So maybe babies should expect those yellow balls to squeak as well.

因此，婴儿也许会觉得 黄色的球也能捏响。

Now, those yellow balls have funny sticks on the end,

这些黄色的球在尾端有一根棍子，

so babies could do other things with them if they wanted to.

因此婴儿还可以对它做其他动作。

They could pound them or whack them.

比如说打它或者掰它。

But let's see what the baby does.

让我们来看婴儿会怎么做。

(Video) Hyowon Gweon: See this? (Ball squeaks)

（视频）权孝媛：看到没？ （球被捏响）

Did you see that? (Ball squeaks)

听到了吗？ （球被捏响）

Cool.

酷。

See this one?

看到这个球没？

(Ball squeaks)

（球被捏响）

Wow.

哇。

Laura Schulz: Told you. (Laughs)

劳拉·舒尔茨：我就说嘛。（笑）

(Video) HG: See this one? (Ball squeaks)

（视频）权孝媛：看这个。 （球被捏响）

Hey Clara, this one's for you. You can go ahead and play.

克拉拉，这个球给你。 拿着玩吧。

(Laughter)

（笑声）

LS: I don't even have to talk, right?

劳拉·舒尔茨： 我都不必解释了，对吗？

All right, it's nice that babies will generalize properties

好的，婴儿能从蓝色球的特性 推导出黄色球的特性

of blue balls to yellow balls,

这非常棒，

and it's impressive that babies can learn from imitating us,

而且婴儿通过模仿我们 进行学习，令人印象深刻，

but we've known those things about babies for a very long time.

但婴儿的这些特点我们早就知道了。

The really interesting question

真正有意思的是，

is what happens when we show babies exactly the same thing,

我们将上述实验完全重复一遍，

and we can ensure it's exactly the same because we have a secret compartment

我们之所以能保证两次实验完全一样， 是因为装球的箱子有一个隔层，

and we actually pull the balls from there,

实际上我们是从那个隔层里往外拿球，

but this time, all we change is the apparent population

但是这一次， 我们更改了样品库的外观，

from which that evidence was drawn.

也就是说盒子里的球看起来不同了。

This time, we're going to show babies three blue balls

这一次，我们还是 给婴儿看三个蓝色的球，

pulled out of a box of mostly yellow balls,

但是装球的箱子里几乎全是黄色的球，

and guess what?

猜猜结果会怎样？

You [probably won't] randomly draw three blue balls in a row

从几乎全是黄色球的箱子里

out of a box of mostly yellow balls.

连续拿出三个蓝色的球， 也许很难。

That is not plausibly randomly sampled evidence.

这不是令人信服的随机取样。

That evidence suggests that maybe Hyowon was deliberately sampling the blue balls.

也许孝媛是故意选的蓝色的球。

Maybe there's something special about the blue balls.

也许蓝色的球有些特别之处。

Maybe only the blue balls squeak.

也许只有蓝色的球能捏响。

Let's see what the baby does.

我们来看婴儿会怎么做。

(Video) HG: See this? (Ball squeaks)

（视频）权孝媛：看到了吗？ （球被捏响）

See this toy? (Ball squeaks)

再看这个。 （球被捏响）

Oh, that was cool. See? (Ball squeaks)

哦，太酷了。看！ （球被捏响）

Now this one's for you to play. You can go ahead and play.

这个是给你的。 拿去玩吧。

(Fussing) (Laughter)

（不耐烦） （笑声）

LS: So you just saw two 15-month-old babies

劳拉·舒尔茨：2个15个月大的婴儿

do entirely different things

仅仅基于他们观察到的取样几率

based only on the probability of the sample they observed.

做出了完全不同的反应。

Let me show you the experimental results.

让我们来看一下实验结果。

On the vertical axis, you'll see the percentage of babies

在纵轴上，你看到的是在不同情况下

who squeezed the ball in each condition,

会去捏球的婴儿的百分比，

and as you'll see, babies are much more likely to generalize the evidence

如图表所示，当婴儿认为取样具有代表性

when it's plausibly representative of the population

而不是特意选取的时候

than when the evidence is clearly cherry-picked.

他们有更高几率去捏黄色的球。

And this leads to a fun prediction:

这个结果能导致一个有趣的推测：

Suppose you pulled just one blue ball out of the mostly yellow box.

假设你从几乎全是黄色球的箱子里 拿出一个蓝色球。

You [probably won't] pull three blue balls in a row at random out of a yellow box,

你也许很难从很多黄球的箱子里 连续拿出三个蓝色球，

but you could randomly sample just one blue ball.

但随机拿出一个还是有可能的。

That's not an improbable sample.

这不是一个小概率事件。

And if you could reach into a box at random

如果你从箱子里随机抽出一个东西，

and pull out something that squeaks, maybe everything in the box squeaks.

而这个东西能捏响， 那可能箱子里所有东西都能捏响。

So even though babies are going to see much less evidence for squeaking,

因此，尽管婴儿们在接下来的 “只拿一个球”的实验中，

and have many fewer actions to imitate

看到的证据更少，

in this one ball condition than in the condition you just saw,

可模仿的动作也更少，

we predicted that babies themselves would squeeze more,

但我们推测婴儿们捏球的几率会升高，

and that's exactly what we found.

结果正是如此。

So 15-month-old babies, in this respect, like scientists,

15个月大的婴儿，在这个实验中， 跟科学家一样，

care whether evidence is randomly sampled or not,

十分看重取样是否真正随机，

and they use this to develop expectations about the world:

他们通过这种方法 来发展对世界的预判：

what squeaks and what doesn't,

什么能捏响，什么不能，

what to explore and what to ignore.

什么值得探究，什么可以忽略。

Let me show you another example now,

下面我们来看另一个实验，

this time about a problem of causal reasoning.

关于因果推论的实验。

And it starts with a problem of confounded evidence

这个实验源于一个让我们所有人

that all of us have,

都感到困惑的事实：

which is that we are part of the world.

我们是这个世界的一部分。

And this might not seem like a problem to you, but like most problems,

也许在你看来这根本不算个问题， 但就像许多其他问题一样，

it's only a problem when things go wrong.

只有问题出现时，它才算一个问题。

Take this baby, for instance.

以下面这个婴儿为例。

Things are going wrong for him.

他就碰到了点问题。

He would like to make this toy go, and he can't.

他想把玩具弄响，但是没有成功。

I'll show you a few-second clip.

我给你们放几秒视频。

大体而言，有两种可能：

And there's two possibilities, broadly:

Maybe he's doing something wrong,

也许他玩的方法不对，

or maybe there's something wrong with the toy.

或者玩具坏了。

So in this next experiment,

因此在接下来的实验中，

we're going to give babies just a tiny bit of statistical data

我们会给婴儿少量统计学数据，

supporting one hypothesis over the other,

这些数据能支持某一种可能性，

and we're going to see if babies can use that to make different decisions

我们再看婴儿能否依据这些数据

作出不同的决定。

about what to do.

Here's the setup.

实验是这样的。

Hyowon is going to try to make the toy go and succeed.

孝媛尝试弄响这个玩具，她成功了。

I am then going to try twice and fail both times,

然后我也开始玩，但两次都失败了，

and then Hyowon is going to try again and succeed,

然后孝媛再次尝试，她又成功了，

and this roughly sums up my relationship to my graduate students

也许这是我跟孝媛 在科技水平上差距

in technology across the board.

的很好体现。

But the important point here is it provides a little bit of evidence

这里的关键点在于， 它提供了一点点证据

that the problem isn't with the toy, it's with the person.

证明问题不在于玩具，而在于人。

Some people can make this toy go,

有的人能让玩具发出声音，

and some can't.

有的人则不能。

Now, when the baby gets the toy, he's going to have a choice.

当婴儿拿到玩具之后， 他要做出选择。

His mom is right there,

他妈妈就在旁边，

他可以将玩具交给妈妈， 换一个人，

so he can go ahead and hand off the toy and change the person,

but there's also going to be another toy at the end of that cloth,

同时在那块布的尽头 放着另一个玩具，

and he can pull the cloth towards him and change the toy.

他可以将布拖过来，换一个玩具。

So let's see what the baby does.

我们来看看他会怎么做。

(Video) HG: Two, three. Go! (Music)

（视频）孝媛：二、三，开始！ （音乐）

LS: One, two, three, go!

劳拉·舒尔茨：一、二、三，开始！

Arthur, I'm going to try again. One, two, three, go!

亚瑟，我再试一次。 一、二、三，开始！

YG: Arthur, let me try again, okay?

孝媛：亚瑟，让我再试一次，好吗？

One, two, three, go! (Music)

一、二、三，开始！ （音乐）

Look at that. Remember these toys?

看啊。 记得这些玩具吗？

See these toys? Yeah, I'm going to put this one over here,

看到了吗？我把这个玩具放在这里，

and I'm going to give this one to you.

把这个玩具给你。

You can go ahead and play.

你可以自己玩了。

LS: Okay, Laura, but of course, babies love their mommies.

劳拉·舒尔茨：好吧，劳拉，但是， 小朋友都爱自己的妈妈呀。

Of course babies give toys to their mommies

他玩不转玩具的时候

when they can't make them work.

肯定会把玩具交给妈妈。

So again, the really important question is what happens when we change

那么，让我们看看 把这少量的统计学数据

the statistical data ever so slightly.

进行更换会怎么样。

This time, babies are going to see the toy work and fail in exactly the same order,

这一次，玩具响和不响的顺序跟刚才一样，

but we're changing the distribution of evidence.

但分布情况跟刚才不同。

This time, Hyowon is going to succeed once and fail once, and so am I.

这一次，孝媛会成功一次，失败一次， 我也一样。

And this suggests it doesn't matter who tries this toy, the toy is broken.

那就表明跟人没关系， 是这个玩具有问题。

It doesn't work all the time.

它时好时坏。

Again, the baby's going to have a choice.

同样的，婴儿要做出选择。

她妈妈就在她旁边， 她可以换人来试，

Her mom is right next to her, so she can change the person,

and there's going to be another toy at the end of the cloth.

同样有另一个玩具 放在布的另一头。

Let's watch what she does.

我们来看她会如何选择。

(Video) HG: Two, three, go! (Music)

（视频）孝媛：二、三，开始！ （音乐）

Let me try one more time. One, two, three, go!

我再试一次。 一、二、三，开始！

Hmm.

嗯？

LS: Let me try, Clara.

劳拉·舒尔茨：克拉拉，让我试一下吧。

One, two, three, go!

一、二、三，开始！

Hmm, let me try again.

嗯，我再试一次。

One, two, three, go! (Music)

一、二、三，开始！ （音乐）

HG: I'm going to put this one over here,

孝媛：我把这个放在这边，

and I'm going to give this one to you.

把这个给你。

You can go ahead and play.

你可以玩了。

(Applause)

（掌声）

LS: Let me show you the experimental results.

劳拉·舒尔茨：我们来看看实验结果。

On the vertical axis, you'll see the distribution

在纵轴上，显示的是

of children's choices in each condition,

在不同情况下婴儿所做选择的比例，

and you'll see that the distribution of the choices children make

我们可以看到，婴儿们做出的选择

depends on the evidence they observe.

跟他们观察到的证据有关。

So in the second year of life,

因此，在出生后的第二年，

babies can use a tiny bit of statistical data

婴儿已经可以利用少量统计数据

to decide between two fundamentally different strategies

来决定如何从两种不同的 基本策略中做出选择

for acting in the world:

从而在这个世界生存：

asking for help and exploring.

求助和探索。

I've just shown you two laboratory experiments

我刚刚向大家展示的两个实验

out of literally hundreds in the field that make similar points,

是从几百个类似实验中挑选出来的， 它们得出了相似的结论，

because the really critical point

因为真正重要的一点是

is that children's ability to make rich inferences from sparse data

孩子们从很少的数据中 推导出丰富结果的能力

underlies all the species-specific cultural learning that we do.

构成了我们研究 物种特异性文化的基础。

Children learn about new tools from just a few examples.

孩子能通过几个示范 就掌握工具的用法。

They learn new causal relationships from just a few examples.

能通过几个例子 就掌握新的因果关系。

They even learn new words, in this case in American Sign Language.

他们甚至能学会新的词语， 这里我指的是美国手语。

I want to close with just two points.

我想用两个观点来结束演讲。

If you've been following my world, the field of brain and cognitive sciences,

如果在过去几年， 你一直在关注我们的领域，

for the past few years,

关注大脑和认知科学，

three big ideas will have come to your attention.

那么你一定注意到了这三个观点。

The first is that this is the era of the brain.

首先，现在是大脑的时代。

And indeed, there have been staggering discoveries in neuroscience:

实际上，神经系统科学 已经取得了不错的进展：

localizing functionally specialized regions of cortex,

确定大脑皮层各区域的作用，

turning mouse brains transparent,

让小白鼠的大脑透明化，

activating neurons with light.

利用光线触发神经元（活动）。

A second big idea

第二个大的观点是

is that this is the era of big data and machine learning,

现在是大数据和机器学习的时代，

and machine learning promises to revolutionize our understanding

机器学习预示了我们对事物 的理解将发生革命性的变化，

of everything from social networks to epidemiology.

无论是对社交网络还是流行病学。

And maybe, as it tackles problems of scene understanding

也许，随着它被用于场景理解

and natural language processing,

和自然语言处理，

to tell us something about human cognition.

能帮助我们更好地研究人类认知。

And the final big idea you'll have heard

最后一个你可能注意到的观点是

is that maybe it's a good idea we're going to know so much about brains

我们能深入了解大脑， 能深入运用大数据，

and have so much access to big data,

是一件非常好的事情，

because left to our own devices,

因为人类天性随意，

humans are fallible, we take shortcuts,

我们容易犯错，喜欢走捷径，

we err, we make mistakes,

我们闯祸，我们惹麻烦，

we're biased, and in innumerable ways,

我们心存偏见， 而且从许多方面来讲，

we get the world wrong.

我们会错误理解这个世界。

I think these are all important stories,

我认为这些书都很重要，

and they have a lot to tell us about what it means to be human,

能帮我们理解身为人类意味着什么，

but I want you to note that today I told you a very different story.

但我想强调的是， 今天我讲的是一个完全不同的故事。

It's a story about minds and not brains,

它讲的是思维而不是大脑，

and in particular, it's a story about the kinds of computations

确切的说，是关于人类思维所特有的

that uniquely human minds can perform,

一种计算能力，

which involve rich, structured knowledge and the ability to learn

这种能力让我们学识渊博，

from small amounts of data, the evidence of just a few examples.

帮助我们从少量数据和证据中 进行学习。

从本质上来说， 这是一个关于成长的故事，

And fundamentally, it's a story about how starting as very small children

and continuing out all the way to the greatest accomplishments

小孩子如何一天天成长， 取得巨大成就，

of our culture,

为我们的文化做贡献，

we get the world right.

我们对世界的理解又是正确的。

Folks, human minds do not only learn from small amounts of data.

朋友们，人类的思维不光 能从少量数据中进行学习。

Human minds think of altogether new ideas.

人类思维能提炼全新的观点。

Human minds generate research and discovery,

人类思维进行研究和发现，

and human minds generate art and literature and poetry and theater,

人类思维还能创作 艺术、文学、诗歌和戏剧，

and human minds take care of other humans:

人类思维还会关注其他人类：

our old, our young, our sick.

尊老爱幼，救死扶伤。

We even heal them.

让他们痊愈。

In the years to come, we're going to see technological innovations

在未来几年， 我们将看到超出我们想象

beyond anything I can even envision,

的技术创新，

but we are very unlikely

但是我们很可能看不到

to see anything even approximating the computational power of a human child

哪怕仅仅是接近 人类小孩计算能力的技术出现，

in my lifetime or in yours.

可能我们的有生之年都看不到。

If we invest in these most powerful learners and their development,

如果我们对这些最强大的 学习者和他们的发展进行投资，

in babies and children

也就是对婴儿和儿童，

and mothers and fathers

对他们的父母，

and caregivers and teachers

对他们的看护和老师，

就像我们对技术、工程和设计 等最强大和优雅的门类

the ways we invest in our other most powerful and elegant forms

of technology, engineering and design,

进行投资一样，

we will not just be dreaming of a better future,

那我们将不仅梦想着更好的未来，

we will be planning for one.

而是按计划在实现它。

Thank you very much.

非常感谢大家。

(Applause)

（掌声）

Chris Anderson: Laura, thank you. I do actually have a question for you.

克里斯·安德森：劳拉，谢谢你。 我有一个问题想问你。

First of all, the research is insane.

首先，这项研究非常棒。

I mean, who would design an experiment like that? (Laughter)

我是说，谁能设计出这样一个实验呢？ （笑声）

I've seen that a couple of times,

我已经看过好几次了，

and I still don't honestly believe that that can truly be happening,

但我仍然不敢相信这是真的，

but other people have done similar experiments; it checks out.

但其他人也做过类似的实验， 真的证明了，

The babies really are that genius.

婴儿们真的都是天才。

LS: You know, they look really impressive in our experiments,

劳拉·舒尔茨：是啊，他们在实验中的表现 真是棒极了，

but think about what they look like in real life, right?

但想象一下他们在生活中 的表现（会更棒），不是吗？

It starts out as a baby.

最开始只是个小东西，

Eighteen months later, it's talking to you,

十八个月后， 他就可以跟你交谈了，

and babies' first words aren't just things like balls and ducks,

婴儿最开始会说的话 不仅仅是球啊鸭子啊这些东西，

they're things like "all gone," which refer to disappearance,

还有包括“不见了”表示消失，

或者“啊—哦”表示下意识的动作。

or "uh-oh," which refer to unintentional actions.

It has to be that powerful.

就是这么神奇。

It has to be much more powerful than anything I showed you.

比我在实验中展示的要神奇得多。

他们能理解整个世界。

They're figuring out the entire world.

一个四岁的小孩几乎能跟你聊任何话题。

A four-year-old can talk to you about almost anything.

（掌声）

(Applause)

CA: And if I understand you right, the other key point you're making is,

克里斯·安德森：如果我没理解错的话， 你想说明的另一个关键点是，

we've been through these years where there's all this talk

多年以来，我们一直认为

of how quirky and buggy our minds are,

人类思维古怪而不正常，

that behavioral economics and the whole theories behind that

行为经济学和它背后的 一整套理论都认为

that we're not rational agents.

人类不是一种理性的生物。

You're really saying that the bigger story is how extraordinary,

而你认为人类思维 是如此卓越，

and there really is genius there that is underappreciated.

如此出色，实际上是被低估了。

LS: One of my favorite quotes in psychology

劳拉·舒尔茨：我最喜欢的 关于心理学的一句话

comes from the social psychologist Solomon Asch,

来自社会心理学家所罗门·阿施，

and he said the fundamental task of psychology is to remove

他说，心理学的基本任务就是

the veil of self-evidence from things.

揭开事物“无证自明”的面纱。

There are orders of magnitude more decisions you make every day

要正确理解世界

that get the world right.

你每天要做出非常之多的决定。

You know about objects and their properties.

你了解物体和它们的属性。

You know them when they're occluded. You know them in the dark.

当有东西挡路的时候你会知道， 即便是在黑暗中。

You can walk through rooms.

你可以穿过房间。

You can figure out what other people are thinking. You can talk to them.

你可以猜到其他人在想什么。 你可以跟他们交谈。

你可以在太空中导航。 你了解数字。

You can navigate space. You know about numbers.

你知道因果关系。 你理解道德推论。

You know causal relationships. You know about moral reasoning.

这些事情做起来不费功夫， 因此我们注意不到，

You do this effortlessly, so we don't see it,

但我们就是这样来正确理解世界的， 这是一种非凡的，

but that is how we get the world right, and it's a remarkable

但非常难以理解的成就。

and very difficult-to-understand accomplishment.

克里斯·安德森：我猜观众中间

CA: I suspect there are people in the audience who have

一定有技术加速理论的支持者，

this view of accelerating technological power

他们可能不认同你的观点， 就是有生之年都看不到

who might dispute your statement that never in our lifetimes

计算机的智能 达到一个三岁孩子的水平，

will a computer do what a three-year-old child can do,

但毫无争议的是，无论如何，

but what's clear is that in any scenario,

our machines have so much to learn from our toddlers.

从蹒跚学步的儿童身上 机器可以学到很多很多。

LS: I think so. You'll have some machine learning folks up here.

劳拉·舒尔茨：的确是。观众中 有从事机器学习研究的朋友。

I mean, you should never bet against babies or chimpanzees

我想说，你不能认为婴儿或者黑猩猩

or technology as a matter of practice,

或者技术的差别在于实践，

but it's not just a difference in quantity,

他们之间的差别不在于数量，

it's a difference in kind.

而在于种类。

We have incredibly powerful computers,

我们现在有非常强大的计算机，

and they do do amazingly sophisticated things,

它们能完成非常精确的任务，

often with very big amounts of data.

处理海量的数据。

Human minds do, I think, something quite different,

但人类思维的运作方式完全不同，

and I think it's the structured, hierarchical nature of human knowledge

我认为研究人类知识 在结构和层次方面的属性

that remains a real challenge.

仍是一项巨大的挑战。

CA: Laura Schulz, wonderful food for thought. Thank you so much.

克里斯·安德森：劳拉·舒尔茨， 带来了美妙的精神食粮。非常感谢。

LS: Thank you. (Applause)

劳拉·舒尔茨：谢谢。 （掌声）